Int. J. Mol. Sci. 2002, 3, 755-763
International Journal of
Molecular Sciences ISSN 1422-0067 © 2002 by MDPI www.mdpi.org/ijms/
Reverse Phase Thin Layer Chromatography of Aminoalkanethiosulfuric Acids, Mercaptoalkanamines and Aminoalkyl Disulfides Maria das Graças Cardoso1, David Lee Nelson2*, Antônia Tavares do Amaral3, Custódio Donizete dos Santos1, Alcilene de Abreu Pereira1, Ana Cláudia Barneche de Oliveira1 1
Departamento de Química, Universidade Federal de Lavras, CP 37, CEP 37200-000 Lavras, MG, Brazil 2 Departamento de Alimentos, Faculdade de Farmácia, Universidade Federal de Minas Gerais, 30180112 Belo Horizonte, MG, Brazil. 3 Departamento de Química Teórica, Instituto de Química, Universidade de São Paulo, São Paulo, SP, Brazil Tel.: 55-31-9981 8792; Fax: 55-31-3339 7666; E-mail:
[email protected]. *Author to whom correspondence should be addressed. Received 30 April 2001 / Accepted 20 March 2002 / Published 31 July 2002
Abstract: The experimental Rm values for a series of aminoalkanethiosulfuric acids, mercaptoalkanamines and aminoalkyl disulfides were determined by reverse phase thin layer chromatography. The Rm values were determined for various concentrations of methanol:water and the correlation obtained was extrapolated to 100% water. These values permitted the calculation of the log P values for each substance. The log P values obtained by this method were compared with those obtained using the shake-flask method and by theoretical calculations utilizing fragment constants. Keywords: aminoalkanethiosulfates, mercaptoalkanamines, aminoalkyl disulfides, thin layer chromatography, log P.
Introduction Chromatographic methods have frequently been applied in the determination of relative partition coefficients (P) for use in quantitative relationships between chemical structure and biological activity (QSAR) [1]. Reverse phase thin layer chromatography and high performance liquid chromatography are used to evaluate the lipophilicity of a series of organic compounds. These correlations are based on
756
Int. J. Mol. Sci. 2002, 3
the studies of Dearden [2], Consdem [3] and Martin and Synge [4] on the relationship between chromatographic distribution and the partition coefficient. The correlation which involves Rm is defined by equation 1 [5]: Rm = log (1/Rf - 1)
(1)
They demonstrated experimentally that the relationship predicted by Martin and Synge [4] was indicated for a number of flavones, anthocyanines and some reference compounds. However, because of the nature of the substituents studied (e.g. hydroxyl groups), the data necessarily were restricted to a narrow range of compounds. Therefore, they analyzed the effect of substituents on the Rf value of an unsubstituted compound. According to Tomlinson [6], the partition coefficient can be defined as an equilibrium constant similar to equation 2. P = Ka / Kb
(2)
and ∆Rm is similar to π. Many times the term ∆Rm is ignored and π is expressed as: π = log (1/RfX) – 1
(3)
(1/RfH) – 1 In the study of the structural characteristics of compounds active against the fungus Podosphaera leucotricha using regression analysis, they examined the relationship between the fungicidal activities of a series of alkyldinitrophenols and the ∆Rm values of the substituents. The Rm values were measured in cellulose layers impregnated with 10% ethyl oleate using 60% aqueous ethanol as the mobile phase. A quadratic relationship was found for a series of 4-(1-cyclopentyl-n-akyl)-2,6-dinitrophenols [5] (Equation 4). log Br = 7,583 - 11,815 ∆Rm + 6,434 (∆Rm)2
(4)
Later, Draber et al. [7] studied the herbicidal activity of some triazinones. They demonstrated that the ∆Rm constants of the substituents, obtained from measured Rm values in a reverse phase thin layer chromatography (RPTLC) system with the chromatographic solvent pair — paraffin oil and water/dioxane — together with its π values, correlate satisfactorily with its ability to inhibit electron transport in isolated chloroplasts. Tomlinson [6] determined the Rm values for acids and bases. In this case, Rm was described by the following expression, stipulating that the degree of association in the organic phase can be ignored: Rm = log (1/Rf - 1) + log Ka + [H+] [H+]
(5)
757
Int. J. Mol. Sci. 2002, 3
where Ka is the dissociation constant of the solute and [H+] it is the hydrogen ion concentration in the mobile phase. Iwasa et al. [8] suggested the usefulness of the chromatographic data for QSAR. According to them, the relationship between the partition coefficient and Rm values constitutes an extension of the Collander equation [9, 10]. that correlates the log P values in octanol/water with those obtained in other apolar/polar solvent pairs. Thus, the following equation was formulated: log P = bRm + a
(6)
where a and b are constants. Also, Hulshoff and Perrim [11, 12] emphasized the study of the classification and of the system model, through which the values of Rm can be measured. In measuring of the Rm values by reverse phase chromatographic methods, the systems are generally constituted of sílica gel impregnated with a lipophilic phase (liquid paraffin, mineral oil, noctanol) or chemically bound (C-8, C18) and an aqueous mobile phase of variable composition and polarity. Due to the nature of some of the solutes studied, an increase or decrease in the polarity of the mobile phase is generally necessary to obtain a reasonable migration of the solute and to obtain measurable values of Rf [13]. Boyce and Milborrow [14] have also demonstrated linear relationships between the Rm values and the number of methylene groups in the alkyl chain in a series of N-alkyltriethylamines in an aqueousacetone system. This fact was related to the work of Irsherwood [15] which correlated the water content of the mobile phase with the Rm values of oligosaccharides. Such evidence showed that the composition of the mobile phase can significantly affect Rm and ∆Rm data measured in similar. Biagi and coworkers [16] mention Rm data for values found by extrapolation of the curve obtained for the percent of acetone in an aqueous acetone mixture versus Rm. These values of Rm corresponding to a system containing 100% water can them be used for any QSAR model. Such an extrapolation is generally valid when Rm varies linearly with the fraction of volume of the organic solvent in the aqueous mobile phase, as represented by the equation of Soczewinski – Wachtmeister [17]. Rm = Rmw + bc
(7)
where Rmw is the value of Rm for an mobile phase containing pure water, b is a constant and c represents the concentration of the organic modifier in the mobile phase. To measure log P by the conventional Shake-flask method is difficult and time consuming. It is difficult to determine log P for substances that are not very soluble in water or that cannot be detected by conventional methods. Instead of measuring the log P values by equilibrium methods, partition chromatography data can be used. The main advantages of the chromatographic methods over the direct partition methods for determining the hydrophobicity index consists of the fact that the former are
758
Int. J. Mol. Sci. 2002, 3
simpler to use, are faster and less tedious [6]. In addition, less material is required and these methods can be used to determine the hydrophobicity of drug molecules with very high or very low log P values (such solutes require long equilibration periods in the normal Shake-flask methods), the material doesn't need to be ultra pure; there is no need to make a specific quantitative analysis of the solute and more satisfactory results are obtained. Based on these facts, the objective of the present study was the determination of the chromatographic parameters of three series of sulfur containing alkylamines — aminoalkanethiosulfuric acids (1), aminoalkyl disulfides (2) and mercaptoalkanamines (3) — using a reverse phase system to find satisfactory correlations for the lipophilicity by extrapolation of the values of Rm. These data are compared with those obtained by the shake flask technique which is used in the determination of the partition coefficient.
R'
R' NH
NH SSO3 H
R
R
R' hexyl hexyl hexyl hexyl hexyl
Comp. 1f1g1h1i1j-
R isobutyl butyl isobutyl butyl sec-butyl
NH S
S
R
N-alkyl-2,2'-dithiobis(1-alkanamines) (2a-c)
Aminoalkanethiosulfuric Acids (1a-j) Comp. R 1a- isopropyl 1b- sec-butyl 1c- butyl 1d- propyl 1e- cycloexyl
R'
R' hexyl pentyl pentyl ethyl ethyl
Comp. R 2a- cyclohexyl 2b- butyl propyl 2c-
R' methyl methyl methyl
R' NH R
SH
N SH
N-alkyl-2-mercapto-1-alkanamines (3a-g) Comp. 3a3b3c3d-
R butyl isopropyl propyl butyl
R' Comp. R R' methyl 3e- cycloexyl methyl methyl 3f- butyl H methyl 3g- cycloexyl H ethyl
N,N-Diisopropyl-2-mercapto-1-propanamine (4)
759
Int. J. Mol. Sci. 2002, 3 Materials and Methods
The compounds studied were synthesized in this laboratory by previously established methods [1821] and their structures were established by physical and spectroscopic techniques (infrared, nuclear magnetic resonance and mass spectrometry).
Determination of the Chromatographic Parameter Rmwol (the Rm value for pure water as the eluent) Glass plates (10 x 10 cm) were coated with a 0.25 mm layer of silica gel DGF-254 (Merck). The plates were activated for 5 h at 110 oC, and, after cooling in a dessicator, were placed in a closed chromatography chamber in contact with a solution 5% of octanol in chloroform for 18 h to impregnate them with the reverse phase [13]. Individual solutions containing 0.3 mg/ml of each of the 21 derivatives in methanol were applied to the chromatographic plates. The mobile phase was composed of binary mixtures of 0.2 M ammonium acetate buffer (pH 6,0) and methanol in the following proportions 50:50; 60:40; 70:30; 80:20; 90:10. The plates were developed at 25 oC. After the mobile phase had migrated for a distance of 8.5 cm, the plates were removed from the chambers and the solvent was evaporated. The spots were visualized with iodine vapor and the Rf values were determined. From the average of the results from three plates. From the Rf, values, the Rm values could be calculated from the following equation: Rm = log (1/Rf-1)
(8)
Results and Discussion To obtain the RmWOL values, the chromatographic parameter for the pure aqueous eluent the linear correlation of the Rm values versus the percent of organic solvent in the eluent was extrapolated to the 0% point. To guarantee the experimental efficiency in the sense of assuring the reproducibility of the results the Rf values obtained in the absence and in the presence of octanol were compared. These values are shown in Table 1. The RmWOL values obtained with varying proportions of methanol in the eluent can be observed in Table 2. The correlation coefficient shown in Table 2 indicates that this is indeed a linear correlation, thus assuring that extrapolation will furnish the RmWOL value for the 100% aqueous eluent (Equation 9). RmWOL = a (% methanol) + b
(9)
760
Int. J. Mol. Sci. 2002, 3
Table 1. Comparison of Rf values of 2-(N-alkylamino)-1-hexanethiosulfuric acids (R´ = hexyl) obtained on plates coated with silica gel with and without being impregnated with octanol. (Eluent: methanol: ammonium acetate buffer = 40:60, pH = 6.0). Substance
1a 1b 1c 1d 1e 1f
R isopropyl sec-butyl butyl propyl cicloexyl isobutyl
Without Octanol 0.655 0.614 0.659 0.683 0.603 0.694
With Octanol 0.555 0.477 0.411 0.422 0.242 0.320
The results obtained from reverse phase thin layer chromatography showed that the tested compounds did not migrate when the mobile phase was only composed of ammonium acetate buffer. The addition of the methanol was necessary to obtain longer migrations and, therefore, more reliable Rm values. The comparison of the Rm values versus the composition of the mobile phase showed a linear relationship. An increases of 10% of methanol in the eluent resulted in a decrease of 2.6 in the Rm values for substance 1a. The experimental data for the mobile phase containing 5% methanol were not used because, at this concentration, several compounds showed very small migrations, presenting, therefore, Rm values outside the limits of maximum precision. The RmWOL values are shown in Tables 3 and 4, together with the log P values in the series of the compounds discussed above. Table 2. Rm values with varying proportions of ammonium acetate buffer (pH = 6.0) and methanol Proportions of ammonium acetate buffer:methanol Substance 50:50 60:40 70:30 80:20 90:10 r* -0.158 -0.096 0.159 0.301 0.720 -0.932 1a 0.000 0.161 0.493 0.673 1.043 -0.990 1b 1c -0.036 0.156 0.505 0.624 0.861 -0.905 -0.099 0.136 0.549 0.623 1.194 -0.952 1d 0.133 0.364 0.547 0.719 1.130 -0.980 1e 0.234 0.327 0.405 0.550 0.720 -0.986 1f -0.434 -0.154 0.020 0.235 0.347 -0.980 1g -0.292 -0.076 -0.076 0.039 0.239 -0.888 1h -0.299 -0.286 -0.286 0.091 0.188 -0.805 1i -0.117 0.012 0.010 0.029 -0.124 -0.914 1j 0.610 0.632 0.726 0.733 0.784 -0.937 2a 0.659 0.710 0.833 0.869 0.939 -0.937 2b 0.739 0.743 0.954 1.026 1.091 -0.925 2c 0.282 0.474 0.796 0.969 1.150 -0.987 3a -0.110 0.034 0.582 0.869 1.007 -0.980 3b 0.120 0.513 0.733 0.921 1.179 -0.962 3c 0.726 0.845 1.020 1.136 1.143 -0.936 3d 0.094 0.133 0.206 0.383 0.598 -0.960 3e 0.573 0.580 0.784 0.903 1.015 -0.953 3f 0.781 0.788 0.821 0.916 0.949 -0.912 3g -0.578 -0.534 -0.412 -0.278 -0.141 -0.974 4
761
Int. J. Mol. Sci. 2002, 3
Table 3. RmWOL, log PAp* and log Pcalc** values for the alkylaminoalkanethiosulfuric acids [22-24]. Substance
RmWOL
log PAp
log PCalc Hansch e Leo Nys e Rekker -0.44 -0.44 -0.58 -0.47 0.03 0.08 0.04 0.04 -0.18 0.04 0.35 0.77 -0.51 -0.44 -0.58 -0.04 -2.06 -2.02 -2.27 -2.05 -2.15 -2.28
1.31 -0.30 1a 0.83 -0.38 1b 1.18 -0.34 1c 0.80 -0.46 1d 1.25 -0.38 1e 1.27 -0.15 1f 0.58 -0.72 1g 0.31 -0.54 1h 0.28 -1.85 1i 1j 0.16 -2.00 -0.04 -2.10 4 *Determined by the shake-flask method. **Calculated from fragmental constants by the methods of Hansch and Leo and of Nys and Rekker
Table 4. RmWOL,, log PAp* and log Pcalc** values for the mercaptoalkanamines and aminoalkyl disulfides [22-24]. Substance RmWOL log PAp log Pcal Hansch e Leo Nys e Rekker 1.20 -1.10 -1.01 2.12 2a 1.10 -0.26 -0.07 3.75 2b 0.83 -0.14 -0.11 4.47 2c 1.44 -1.64 -0.16 2.03 3a 1.39 -1.76 -0.30 2.01 3b 1.13 -1.57 -0.09 2.08 3c 1.40 0.14 0.31 2.57 3d 1.31 -0.39 -0.26 3.09 3e -0.17 2.67 0.99 -1.65 3f 0.66 -0.52 0.23 3.25 3g *Determined by the shake-flask method. **Calculated from fragmental constants by the methods of Hansch and Leo and of Nys and Rekker
The analysis of the quantitative structure-activity relationship of the series of aminoalkanethiosulfuric acids, mercaptoalkanamines and aminoalkyl disulfides lead to the following equations (calculated and observed values) [23].
Conclusions The chromatographic values of the three series of sulfur-containing amines – aminoalkanethiosulfuric acids, mecaptoalkanamines and aminoalkyl disulfides were determined by reverse phase thin layer
Int. J. Mol. Sci. 2002, 3
762
chromatography. A satisfactory correlation with other methods was found when the extrapolated values of Rm were used. Thus, this technique can be used in place of more tedious methods for these classes of compounds.
Acknowledgements Financial support was provided by the Fundação de Apoio à Pesquisa do Estado de Minas Gerais (FAPEMIG) and the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq). Drs. D. L. Nelson, M. G. Cardoso, A. T. do Amaral and C. D. dos Santos were recipients of research fellowships from the CNPq.
References and Notes 1. Bate-Smith, E. C.; Westall, R. G. Biochem. Biophys. Acta. 1950, 4, 427-440. 2. Dearden, J. C.; Mays, P. K. J. Pharm. Pharmacol. 1985, 61, 203. 3. Consdem, R.; Gordom, A. H.; Martim, A. J. P. Biochem. J. 1944, 35, 224-232. 4. Martin, A. J. P.; Synge, R. L. M. Biochem. J. 1941, 35, 1358-1368. 5. Clifford, R. D.; Deacon, C. A.; Holgate, E. M. Ann. Appl. Biol. 1969, 64, 131. Apud. Tomlinson, E. J.
Chromatogr. 1975, 113, 1-45. 6. Tomlinson, E. J. Chromatogr. 1975, 113, 1-45. 7. Draber, W., Buchel, K. H.; Dickore, K. Proc. Int. Congr. Pestic. Chem. 2nd. 1971, 8. Iwasa, J., Fujita, T.; Hansch, C. J. Med. Chem. 1965, 8, 150-153. 9. Collander, R. Acta. Chim. Scand. 1951, 5, 774-780. 10. Collander, R. Physiol. Plant. 1954, 7, 420-445. 11. Hulshoff, A.; Perrim, J. H. J. Chromatogr. 1976, 120, 65-80. 12. Hulshoff, A.; Perrim, J. H. J. Chromatogr. 1976, 129, 263-276. 13. Corrêa, C.J.C. Correlações quantitativas entre estrutura e atividade antiinflamatória de derivados
aminopirazólicos. Rio de Janeiro, Instituto de Química da UFRJ, 1992. 85p. (Masters Thesis em Science – Organic Chemistry). 14. Boyce, C. B. C., Milborrow, B. V. Nature. 1965, 208, 537-539. 15. Isherwood, F. A.; Hanes, C. S. Biochem. J. 1953, 55, 824-830. 16. Biagi, G. L.; Barbaro, A. M.; Gamba, M. F.; Guerra, M. C. J. Chromatogr.1969, 41, 371-379. 17. Soczewinski, E.; Wachtmeister, C. A. J. Chromatogr. 1962, 7, 311-320. 18. Vieira Filho, S. A.; Nelson, D. L.; Piló-Veloso, D. Anais Acad. Bras. de Ciências 1986, 58, 547551. 19. Penido, M. L. O.; Nelson, D.L.; Piló-Veloso, D. J. Braz. Chem. Soc. 1990, 1, 35-40,
Int. J. Mol. Sci. 2002, 3
763
20. Penido, M. L. O. ; Nelson, D. L., Vieira, L. Q.; Coelho, P. M. Z. Mem. Inst. Oswaldo Cruz 1994,
89, 595-602 21. Moreira, L. S. A.; Piló-Veloso, D.; Nelson, D. L. Química Nova 2000, 23, 447-452. 22. Leo, A.; Hansch, C., Ed. Comprehensive Medicinal Chemistry, Oxford: Pergamon Press, 1990, 4, 295. 23. Cardoso, M. G.; Nelson, D. L.; Santos, C. D.; Amaral, A. T.; Souza, P. E.; Oliveira, R. M. Ciência e
Agrotecnologia 1997, 21, 465-468. 24. Nys, G.G.; Rekker, R.F. Eur. J. Med. Chem. 1974, 9, 361-375.
© 2002 by MDPI (http://www.mdpi.org).